of 39 ACS Paragon Plus Environment Langmuir 1 ... - ACS Publications

Subscriber access provided by La Trobe University Library

Article

Interfacial Properties and Mechanisms Dominating Gas Hydrate Cohesion and Adhesion in Liquid and Vapor Hydrocarbon Phases Sijia Hu, and Carolyn A. Koh Langmuir, Just Accepted Manuscript • DOI: 10.1021/acs.langmuir.7b02676 • Publication Date (Web): 18 Sep 2017 Downloaded from http://pubs.acs.org on October 1, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Gas hydrate interparticle force increases with increasing contact time. 246x197mm (144 x 144 DPI)

ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Interfacial Properties and Mechanisms Dominating Gas Hydrate Cohesion and Adhesion in Liquid and Vapor Hydrocarbon Phases Sijia Hu1, Carolyn A. Koh1,* 1. Center for Hydrate Research, Chemical & Biological Engineering Department, Colorado School of Mines, Golden, CO 80401

ABSTRACT The interfacial properties and mechanisms of gas hydrate systems play a major role in controlling their interparticle and surface interactions, which is desirable for nearly all energy applications of clathrate hydrates. In particular, preventing gas hydrate interparticle agglomeration and/or particle-surface deposition is critical to the prevention of gas hydrate blockages during the exploration and transportation of oil and gas subsea flowlines. These agglomeration and deposition processes are dominated by particleparticle cohesive forces and particle-surface adhesive force. In this study, we present the first direct measurements on the cohesive and adhesive force studies of CH4/C2H6 gas hydrate in a liquid hydrocarbon-dominated system utilizing a high pressure micromechanical force (HP-MMF) apparatus. A CH4/C2H6 gas mixture was used as the gas hydrate former in the model liquid hydrocarbon phase. For the cohesive force 1 ACS Paragon Plus Environment

Page 2 of 39

Page 3 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

baseline test, it was found that the addition of liquid hydrocarbon changed the interfacial tension and contact angle of water in liquid hydrocarbon compared to water in the gas phase, resulting in a force of 23.5 ± 2.5mN m-1 at 3.45 MPa, 274 K and 2-hour annealing time period in which hydrate shell growth occurs. It was observed that the cohesive force was inversely proportional to the annealing time, whereas the force increased with increasing the contact time. For longer contact time (> 12 hours), the force could not be measured because the two hydrate particles adhered permanently to form one large particle. The particle-surface adhesive force in the model liquid hydrocarbon was measured to be 5.3 ± 1.1 mN m-1 at the same experimental condition. Finally, with 1-hour contact time, the hydrate particle and the carbon steel (CS) surface were sintered together and the force was higher than what could be measured by the current apparatus. A possible mechanism is presented in this paper to describe the effect of contact time on particle-particle cohesive force based on the capillary liquid bridge model. A model adapted from the capillary liquid bridge equation has been used to predict the particleparticle cohesive force as a function of contact time, showing close agreement with the experimental data. By comparing the cohesive forces results from gas hydrates both gas and liquid-bulk phases, the surface free energy of a hydrate particle was calculated and found to dominate the changes in the interaction forces with different continuous bulk phases.

2 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

INTRODUCTION Gas hydrates belong to a class of compounds, commonly known as clathrates. Clathrate hydrates are crystalline compounds with molecular cage structures, where the host water molecules form a network of hydrogen-bonded water cages that encapsulate the guest molecules.1-3 The guest molecules, such as methane (CH4), ethane (C2H6), propane (C3H8), or their mixtures, can fit within the hydrogen-bonded water cages typically with a maximum of one guest molecule per water cage. Gas hydrates usually form at high pressure and relatively low temperature conditions (e.g., 2 MPa and 277 K). In general, gas hydrates are classified by three crystal structures: cubic structure I (sI), cubic structure II (sII), and hexagonal structure (sH). The crystal structure of gas hydrates is mainly controlled by the size of the guest molecules. Since all common hydrate structures consist about 85 mole% water, the physical properties of ice and hydrates are quite similar.3 Gas hydrates can be formed in subsea flowlines during the exploration and transportation of oil and gas, due to the prevalent pressure and temperature conditions.4 Interfacial thermodynamics, including the changes to chemical potential, interfacial tension, Gibbs free energy and entropy during gas hydrate nucleation or growth are of great interest, and are the key areas requiring better understanding of the interfacial properties for gas hydrates.5 Due to the relatively large surface area produced in the hydrate system, interfacial tension can contribute substantially to the total free energy.5-7 When gas hydrate particles are present in different bulk phases (gas, liquid hydrocarbon, or even water), the particle-particle and particle-surface interactions change based on the changes to the interfacial tension8, contact angle, as well as the Gibbs free energy. 3 ACS Paragon Plus Environment

Page 4 of 39

Page 5 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

In typical flowlines, especially when water is present, gas hydrate particles can agglomerate and deposit on the flowline walls during both fluid flow or shut-in/restart conditions, resulting in a reduction in the flow cross-sectional area and pressure buildup.3, 9

Figure 1 illustrates four key concepts for gas hydrate plug formation, including major

phenomena that occur in the flowlines while water, oil and gas are present at the same time: (1) water entrainment in oil; (2) gas hydrates nucleation and growth; (3) agglomeration of gas hydrate particles; (4) plug formation due to jamming, deposition, bedding and agglomeration.

Figure 1. Conceptual picture of gas hydrate plugging formation in oil and gas flowlines. (reproduced from Aman et al.5 ) Due to the interests in the mechanism of gas hydrate agglomeration and deposition, previous studies have been performed to investigate clathrate hydrate particle-surface and particle-particle interactions. Yang et al.11 studied the particle-particle cohesive force of both ice and clathrate hydrates formed by tetrahydrofuran (THF) at atmosphere pressure and temperatures of 263-275 K. They also measured the temperature dependence and found that the cohesive force is proportional to the temperature. The force between particles can be explained by capillary liquid bridging (Figure 2) and surface roughness. 4 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2 illustrates the particle-surface (top) and particle-particle (bottom) interactions with the capillary liquid bridge (CLB), further investigations and predictions related to this CLB theory will be discussed in later sections.

Figure 2. Schematic of the particle-surface (top) and particle-particle (bottom) interaction with the capillary liquid bridge, showing relevant parameters: embracing angle (α), capillary bridge (χ), contact angle (θp), immersion depth (dsp/sp(H,V)), distance between particles (H), bridge radius of curvature (r), particle radius (R) and the plane symmetry (A-A) (redrawn from Wang12 and Rabinovich et al.13).


Page 6 of 39

Page 7 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Aspenes and co-workers14 investigated the particle-surface interaction using cyclopentane (CyC5) as the clathrate hydrate former. The results showed that the adhesive force was dependent on the surface material, due to their different surface free energy. They also found the force was strongly dependent on the presence of water. Dieker15 found that CyC5 was a better clathrate hydrate model former than THF, because THF can also form an ice phase (since the experiments were typically performed below the ice point to facilitate a sufficient driving force for THF hydrate to form). Dieker15 also measured the force in a system with a small amount of natural oil, finding that natural oils with a higher content of asphaltenes and acids may exhibit non-plugging tendencies. Smith et al.16 tested the effect of surface energy and wettability using THF hydrate particles, and demonstrated a reduction in adhesion strength by treated coatings. Aman et al.5,

10, 17

measured both the cohesive and adhesive forces as a function of

contact time, contact force and temperature, using the CyC5 model hydrate former, which is stable at ambient pressure conditions. A hydrate interparticle force model was also presented, including capillary and sintering contributions. [Note: sintering is a reactive mechanism that refers to the conversion of the liquid capillary bridge at the contact point between hydrate particles to solid hydrate.10] In addition, they presented a cohesive force measurement in a nitrogen gas phase, showing the same magnitude with the liquid phase measurement. Brown and co-workers9 measured the effect of annealing time (the period during hydrate shell growth) and surfactants using CyC5 hydrate particles. [Note: annealing refers to the conversion of a liquid droplet to a solid hydrate particle.] They found that the cohesive force was inversely proportional to the annealing time, and hydrate shell strength was reduced when a surfactant was present in the system. They


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

also evaluated the performance of modified coatings, showing a 96% reduction for the liquid phase measurements, and about a 40% reduction for the gas phase when comparing a corroded surface to a omniphobic surface.18 Jung et al.19 measured both the gas and THF hydrate tensile strength by applying an external pull-out force, with a measured value of 0.20 ± 0.03MPa. Song et al.20 presented direct contact force measurements between CyC5 hydrate and water in different liquid baths, using a microbalance and zaxis stage, finding that the contact force decreases with decreasing interfacial tension. Lee and co-workers21 used the same apparatus to investigate the effect of salts. It was found that the salts can significantly reduce the adhesive force for a system containing a surfactant, whereas salts did not profoundly influence the force in a surfactant-free system. Recently, a high pressure micromechanical force apparatus was used to measure gas hydrate particle interactions in the vapor phase.12, 22 The cohesive force for CH4/C2H6 hydrates was found to be ~33mN m-1 22 and Wang and co-workers12 showed comparable results of 35 mN m-1 for these gas hydrate particles. The effect of corrosion and the concentration of salts were investigated by Wang et al.12 and it was found the contact force was proportional to the level of corrosion, but inversely proportional to the concentration of salts. Previous studies mainly focused on contact force measurements using liquid hydrate formers stable at atmospheric pressure, such as THF and CyC5, which indicated a significant gap between model clathrate hydrates and gas hydrates under more realistic conditions. Although recent work presented measurements using CH4/C2H6 hydrates in a gas-dominated system, the comparison with gas hydrates in a liquid hydrocarbon-


Page 8 of 39

Page 9 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

dominated system remained an outstanding question. The interfacial tension and surface energy will be different due to the presence of the liquid phase under pressure. In this study, a high-pressure micromechanical force (HP-MMF) apparatus was adapted and applied to directly measure gas hydrate adhesive/cohesive forces under low temperature and high pressure conditions in the presence of a liquid hydrocarbon phase. We introduce a new method to perform these high pressure interparticle measurements in a model liquid hydrocarbon phase, which enables gas hydrate interactions to be compared in both gas-dominated and liquid hydrocarbon-dominated environments. Cohesive and adhesive force measurements between CH4/C2H6 gas hydrates and a bare carbon steel surface in a liquid-dominated system are reported. Adhesive force measurements in a gas-dominated system are also presented at different contact times. These measurements enable us to provide direct investigations in a system with compositions close to the real flowline. The quantitative behavior of gas hydrate interparticle and particle-surface interactions in a liquid hydrocarbon phase have been investigated and compared against the gas-phase and low pressure systems. A model adapted from the capillary liquid bridge equation has been used to predict the particle-particle cohesive force as a function of contact time. The surface free energy of gas hydrate particles is also predicted in both gas and liquid bulk phases.

EXPERIMENTAL APPARATUS AND METHODS In this study, an optimized HP-MMF apparatus was utilized, which enables cohesive and adhesive force measurements to be performed with gas hydrate formers, in this case a CH4/C2H6 (74.7 mol.%/25.3 mol.%) gas mixture. The forces were calculated using


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Hooke’s law at the desired experimental parameters, such as contact time, annealing time and operating temperature and pressure. Figure 3 shows the schematic of the modified high pressure MMF system including the testing section, aluminum cell, stirrer and stir bar (VWR; Model 230), feed-throughs, nanomanipulator (manufactured by Klocke Nanotechnik), pressure transducer (OMEGA, PX309-3KG5V, BSL accuracy of ± 0.08%) and thermocouple (OMEGA, TMQSS-032G6, Uncertainty: greater of 1K or 0.75%). The testing section was made of stainless steel to sustain pressures of up to 10.3 MPa; therefore, a broad spectrum of gas hydrate formers and conditions can be applied and studied. Two polycarbonate windows were mounted at the top and bottom of the cell, which were used to observe the experiment and obtain the results. An aluminum cell containing liquid hydrocarbon was incorporated in the system. The submersible magnetic stirrer and stir bar were utilized to accelerate the gas diffusion in the model liquid hydrocarbon, and therefore reduce the gas hydrate formation time. A 4 cm by 4 cm base was required to hold the apparatus, which has a height of 5.5 cm. The total movement of the nanomanipulator is 2 cm in the x and y directions, and 1 cm in the z direction. The small dimensions of this equipment allowed it to be housed inside a highpressure cell. This nanomanipulator was employed to move the top particle, which can be ice, hydrate, or carbon steel. A stationary cantilever was connected on the lid of the testing section, fixing the position of the bottom particle. The cell temperature was controlled by a chiller which circulates the mixture of ethylene glycol (Sigma-Aldrich, CAS#107-21-1) and deionized water. An Olympus SZ61 microscope equipped with a CCD camera was used to observe the particles, record the measurements and transfer the image to LabView, software used to record the pressure


Page 10 of 39

Page 11 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

and temperature data. For adhesive force measurements, one gas hydrate particle was replaced by a surface sample.

Figure 3. Schematic of the experimental setup of HP-MMF. Cohesive Force Measurements Cohesive force measurements were performed between two gas hydrate particles in this apparatus. To conduct the experiment, several steps were performed. The first step was to create and place water droplets with a diameter of approximately 400 µm on the glass fibers, and the two cantilevers were attached at the lid and nanomanipulator. The gas mixture was saturated with water through a gas saturator before entering the cell to avoid evaporation of the water droplet on the fiber. The water droplet was then quenched in liquid nitrogen to form ice before being placed in the cell. Secondly, the model liquid hydrocarbon was added in the aluminum cell by using a pipette with a controlled volume of 10 mL. The detailed composition of the model hydrocarbon is shown in Table 1. Due to the low rate of gas diffusion in the liquid hydrocarbon phase, a magnetic stirrer was applied to accelerate this process. A stir bar was added into the aluminum cell and then 10 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

the stirrer was placed directly under the cell. It should be noted that although the stirrer can accelerate the rate of gas diffusion (7.5 × 10-4 mol/min without stirring, predicted from Majid23), it may take several hours for hydrate formation in the liquid hydrocarbon phase. Thirdly, the cell was pressurized to 3.45 MPa and the temperature was set to 274 K. After observing the formation of gas hydrates through the microscope, the time was recorded to reach the desired annealing (i.e., hydrate shell growth). It should be noted that the particles were not placed in contact before completing the annealing process. Several annealing times were used in this study to investigate the effect of annealing time. To test the influence of contact time (i.e., to mimic transient operations during hydrocarbon production and transportation), particles were contacted for a desired amount of time until performing the measurements. To obtain accurate results, annealing time (i.e., 2 hours) remained the same for all contact time measurements. Finally, gas hydrate cohesive force measurements were performed in a liquid hydrocarbon phase. A detailed introduction of gas hydrate cohesive force measurements can be found in a previous publication.10 Typically, pull-off measurements were performed to measure the cohesive force between the two gas hydrate particles, the fourstep schematic graph of this measurement is shown in Figure 4: (1) two water droplets were placed on the cantilevers and converted to gas hydrate particles; (2) the top particle was moved down to contact the bottom particle, giving a preload displacement with the contact time of 10s; (3) the top particle was moved up at a constant velocity until the separation of the two particles; (4) the maximum displacement was obtained (∆D in Figure 4). Hooke’s Law was then applied to calculate the cohesive force where it equaled


Page 12 of 39

Page 13 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

to the spring constant of the fiber (previously measured) multiplied by the maximum displacement. In a single experiment, at least 40 pull-off measurements were performed based on the above procedure and the error is defined as the 95% confidence interval of a t-distribution. The values discussed in this work are reported as the average of multiple experimental repeats. Table 1. Composition of the model liquid hydrocarbon. Component C16 C17 C18 C19 C20 C21 C22 C23 C24 C25 C26 C27 C28 C29 C30

Mass Fraction [%] 0.09 1.23 5.22 11.75 16.04 17.04 12.20 6.34 4.23 3.76 3.29 2.66 2.27 1.56 12.34


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. The procedure of the cohesive force measurement. (modified from Aman10 and Wang12): (1) two water droplets were placed on the cantilevers and converted to gas hydrate particles; (2) the top particle was moved down to contact the bottom particle, giving a preload displacement with the contact time of 10s; (3) the top particle was moved up at a constant velocity until the separation of the two particles; (4) the maximum displacement was obtained. Adhesive Force Measurements A carbon steel (CS) surface sample (6 mm × 6 mm × 1.5 mm; supplied by Oceanit Laboratory Inc.) was pre-dried and soaked in liquid nitrogen for approximately 30 seconds. Then it was placed above warm deionized water for condensation of the water vapor onto the surface. A thin layer of water was generated after about 30 seconds. The procedure is shown on Figure 5, where the water layer is deposited on the CS sample. In this work, a thicker water layer was also deposited on the surface due to the interests on investigating the effect of the amount of water on the surface. To deposit a thicker water layer, instead of using the above water condensation method, a water droplet was placed on the surface to form a thick water layer. To conduct the adhesive force measurements, the top gas hydrate particle was replaced by the treated CS sample. The remaining 13 ACS Paragon Plus Environment

Page 14 of 39

Page 15 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

procedure for performing the adhesive force measurements is very similar to that for the cohesive force measurements.

Figure 5. The procedure of carbon steel surface sample preparation: (1) one CS surface was placed on the cantilever; (2) water layer was deposited on the CS surface, converted to gas hydrates when performing the measurements.

RESULTS AND DISCUSSION Cohesive force baseline measurements in model liquid hydrocarbon As deepwater flowlines move to more extreme conditions, such as higher pressure and lower temperature for subsea exploration and oil and gas transportation, gas hydrate nucleation, crystal growth, deposition and agglomeration in a liquid hydrocarbondominated system are important to investigate. Although liquid hydrocarbon is not involved in gas hydrate formation, which requires only water and guest (small gas) molecules, the liquid hydrocarbon, acting as a bulk phase, changes the interfacial thermodynamics, such as interfacial tension and surface energy. As such, the direct measurement of gas hydrate particle-particle and particle-surface interactions are of great interest in this study due to the representation of a liquid hydrocarbon flowline system.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Due to the low diffusion of CH4/C2H6 in the model liquid hydrocarbon, a stirrer was applied to accelerate the gas saturation in the liquid hydrocarbon phase. Figure 6 shows the particles before and after gas hydrate formation in the model liquid hydrocarbon using a CH4/C2H6 gas mixture. These particles were formed at 3.45 MPa and 274 K, giving a subcooling of 9.7 K. Gas hydrate particle cohesive force measurements were conducted after the successful gas hydrate formation in model liquid hydrocarbon. Figure 7 illustrates the baseline test results using an annealing time of 2 hours and contact time of 10 seconds. These parameters were obtained from previous work on a gas-dominated system5, 12; therefore a direct comparison could be performed for these different systems. Five individual experiments are presented in Figure 7 and at least 40 pull-off measurements were performed in each experiment to reduce the fluctuations of single pull-off tests. The cohesive force was calculated using Hooke’s Law and the results shown on the y-axis were the normalized force, equaling to the force divided by the harmonic mean radius of gas hydrate particle pair. The same conditions were applied in all five individual experiments, therefore, the final value of the cohesive force in liquid hydrocarbon was reported as the average of all data points. The error was analyzed using a 95% confidence interval of a t-distribution for all data points. The cohesive force measured in this system was 23.5 ± 2.5 mN m-1 at 3.45 MPa and 274 K with a 2-hour annealing time and 10second contact time.


Page 16 of 39

Page 17 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

300 µm Figure 6. Gas hydrate particles formed using CH4/C2H6 gas mixture at 3.45 MPa and 274 K in the model liquid hydrocarbon with an annealing time of 2 hours (right) compared to before gas hydrate formation with some ice present (left).


Langmuir

40

29.9 27.3 Cohesion Force (mN/m)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 39

30

19.5

20.1

18.9

20

10

0 1

2

3

4

5

Exp No.

Figure 7. Gas hydrate cohesive force baseline measurements using a CH4/C2H6 gas mixture at 3.45 MPa and 274 K in model liquid hydrocarbon with annealing time of 2 hours. Five repeat measurements are shown in the graph. Each measurement contains at least 40 pull-offs. The dashed line represents the overall average value of the cohesive force, after analyzing all data using the method discussed above. The error is defined as the 95% confidence interval of a t-distribution. Effect of annealing time and contact time in liquid hydrocarbon The baseline test revealed the cohesive force at one particular condition. More tests were conducted to investigate the effect of annealing time and contact time as they are critical in hydrocarbon transportation. In a typical flowline, when the pressure and temperature reach the hydrate formation condition, single water droplets may be converted to gas hydrate particles without contacting any other particles (particularly for low water 17 ACS Paragon Plus Environment

Page 19 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

content systems). These gas hydrate particles can anneal (i.e., undergo hydrate shell growth) before agglomerating to form aggregates. When these particles come into contact after annealing (for different shell growth time periods), their interactions are changed compared to the baseline test results. Figure 8 illustrates the dependence of annealing time, which ranges from 2 to 24 hours. As the annealing time increases, the cohesive force decreases from 23.5 ± 2.5 mN m-1 to 4.7 ± 0.7 mN m-1. A possible explanation for this observation was the reduction of water remaining at the surface of the hydrate shell, which indicated more water converted to solid gas hydrates, reducing the quasi-liquid layer and capillary liquid bridge between two gas hydrate particles.24-26 It is likely that the water layer provides some liquid in the capillary bridge that contributes to the cohesion. Therefore, as the annealing time increases, gas hydrate particles in flowlines can become “drier” and less porous, with the unconverted water within the pores between hydrate crystallites in the shell being reduced due to further conversion to gas hydrates. Figure 8 also shows the cohesive forces at atmosphere pressure by using the CyC5 hydrate former. In this system, a similar behavior was observed, but the result ( 3.58 ± 0.29 mN m-1 9) is around one order of magnitude lower than that at high pressure with a liquid hydrocarbon phase. One possible explanation for this observation could be the difference of the hydrate former and bulk which leads to different interfacial mechanisms, another possibility could be the low gas diffusion rate in the model liquid hydrocarbon phase, or a combination of both effects. In the system using CyC5, hydrate particles can be fully annealed in a relatively short time (< 1 hour) due to the liquid CyC5 acting as a hydrate former; in the system using model liquid hydrocarbon, this liquid phase is not a hydrate former and therefore does not participate in the hydrate crystal structure, A


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

CH4/C2H6 (75/25 mole%) gas mixture was applied here to form gas hydrates (sII). Since the gas diffusion occurs at a relatively low rate in liquid hydrocarbon, it takes longer time for gas hydrate former molecules to reach the surface of the water droplet, leading to gas hydrate particles that are not fully converted and a water layer remaining on the surface at shorter annealing time. With increasing annealing time, the unconverted water would be converted to gas hydrate, thereby further reducing the cohesive force. This also explains why at 24 hours, the cohesive force is 4.7 ± 0.7 mN m-1, which is not statistically differentiable from results for CyC5 hydrate.

Figure 8. Annealing (hydrate shell growth) time effect on hydrate particle-particle cohesion in a liquid hydrocarbon phase (■: gas hydrates formed using a CH4/C2H6 gas mixture at 3.45 MPa and 274 K in model liquid hydrocarbon; ●: CyC5 hydrate formed at 19 ACS Paragon Plus Environment

Page 20 of 39

Page 21 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

274 K), with the contact time of 10 seconds. The red dashed line represents the polynomial fit trend line for the results (to help guide the eye). Each point contains at least 40 pull-offs. The error is defined as the 95% confidence interval of a t-distribution. The final value of each point is calculated as the average of multiple repeat experiments. When a sufficient amount of water is present in liquid hydrocarbon flowlines, the gas hydrate particles can form mostly in the bulk phase. These gas hydrate particles can have the opportunity to make contact with each other after the formation of the gas hydrate shell. In other words, these particles may not be annealed for a long period of time, instead, they may contact each other via capillary liquid bridges after varying annealing times. Figure 9 illustrates the effect of contact time from 10 seconds to 24 hours (■ and dashed and solid lines). The contact time here was interpreted as the time that two gas hydrate particles come into contact with one another with a certain pre-load force. The data suggest that the cohesive force is proportional to the contact time. As the contact time was increased to longer than 12 hours, the force exceeded the limitation of the current fiber, which was 150 mN m-1. There are three possible mechanisms for the interactions between gas hydrate particles:8 (1) solid-solid cohesion, where the force is proportional to the product of particle-fluid interfacial tension and area caused by cohesive failure;27-28 (2) capillary liquid bridge cohesion, where the force depends on bridge-fluid interfacial tension and bridge-particle contact angle;13, 29-31 (3) sintering between gas hydrate particles, where the force would be proportional to the product of the solid tensile strength and sintered area.32 Of the possible mechanisms discussed in previous work10, the solid-solid cohesion and the


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

capillary bridge theory were time independent, which indicates that they both were mechanical results of the system and the reaction terms were not included. However, from the data shown in Figure 9 the large force during long contact time was not only caused by the liquid bridge, but also by the reaction of a water bridge to solid gas hydrates. The latter was defined as hydrate sintering, involving gas hydrate growth in the system.10 Before the water layer converted to gas hydrates (hydrate + liquid water), the measured force was dominated by capillarity, whereas after the liquid water layer converted to gas hydrates, and the force was then dominated by hydrate sintering. The predictions of force as a function of contact time are shown in Figure 10 (solid line), and are described in more detail in the later section on the Capillary Bridge Model.


Page 22 of 39

Page 23 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 9. Contact time effect of the cohesive force of gas hydrate particles using a CH4/C2H6 gas mixture at 3.45 MPa and 274 K in model liquid hydrocarbon. The contact time ranges from 10 seconds to 24 hours (86400 seconds). Each point contains at least 40 pull-offs. The error is defined as the 95% confidence interval of a t-distribution. The final value of each point is calculated as the average of multiple repeat experiments. The dashed blue line represents the prediction using the modified dynamic force model by Aman et al.10 Figure 10 illustrates a possible mechanism of gas hydrate agglomeration with long contact time. Gas hydrate particles in the bulk phase interact with each other through a liquid bridge, which will be converted from liquid to solid hydrates after long contact times. This hydrate sintering process results in irreversible interaction. In other words, it would be difficult to separate the sintered particles even with the shear in the flowline. As more gas hydrate particles interact together, the sintering process recurs, and finally leads to large agglomerates comprised of several gas hydrate particles. While these novel results could be of interest to liquid hydrocarbon flowlines, further systematic studies need to be undertaken to understand the contributions of additional variables such as, preload force, shear rate, and the presence of chemicals, such as surfactants.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 10. Conceptual picture of the gas hydrate particle cohesion process, where blue represents a water phase, green a vapor phase, brown a liquid hydrocarbon phase, and white for gas hydrates: (a) gas hydrate particles contact via water capillary bridge interactions; (b) the water capillary bridge is converted to gas hydrate via sintering; (c) more particles contact via the free water in the bulk phase; (d) large hydrate agglomerates formed. Adhesive force measurements in model liquid hydrocarbon In liquid hydrocarbon flowlines, gas hydrate particles can agglomerate together to form large aggregates, and can also deposit on the flowline wall due to the splashing or condensation of water and/or interaction of the particles via capillary bridging. The water coating/condensing on the flowline wall can be easily converted to gas hydrate deposits 23 ACS Paragon Plus Environment

Page 24 of 39

Page 25 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

with sufficient light hydrocarbons present.12 From previous studies and the models for liquid bridging theory between a solid sphere and a flat plate12, 14, 31, water on the flowline wall can have a non-negligible effect on the adhesive force between the surface and gas hydrate particles. As such, the amount of water in this work was carefully controlled to investigate the particle-surface interaction.

Hydrate layer

400 µm Figure 11. Gas hydrate formed on both the carbon steel surface and on the glass fiber using a CH4/C2H6 gas mixture at 3.45 MPa and 274 K. Model liquid hydrocarbon was placed in the bulk phase. Carbon steel surface sample was placed on top of the gas hydrate particle. To yield repeatable and consistent results, a water condensation method was utilized to control the amount of water on the CS samples.12 Figure 11 shows a typical hydratesurface interaction through HP-MMF, where the CS sample was placed on top of the gas hydrate particle, a thin water layer was condensed on the bottom surface of the CS surface sample. By applying the above method, the water amount on the surface was 0.1 ± 0.01 mg mm-2. The measured force between a gas hydrate particle and the surface was 24 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

5.3 ± 1.1 mN m-1 based on 3 repeat experiments and at least 40 pull-offs for each experiment. Similar to the contact time measurements performed for hydrate particle cohesion (see Figure 9), contact time measurements were also performed for hydrate particle-surface adhesion in a model liquid hydrocarbon system. The latter adhesive force measurements showed a three-fold (16.3 ± 2.1 mN/m) increase in the particle-surface adhesive force when increasing the contact time from 10s to 4h (note: this is based on three repeat measurements, but only one pull-off was performed for each experiment). This increase in adhesive force with contact time is significant, but lower than the corresponding cohesive force increase with increasing contact time. Effect of contact time on gas hydrate particle-surface interactions in a CH4/C2H6 vapor phase When the liquid volume is relatively low in the flowline, gas hydrate particles may deposit on the flowline wall in the gas phase due to the water splashing and condensing on the wall. As some hydrate particles deposit on the surface through contact with the water layer and gas hydrate former in the vapor phase to form the first layer of gas hydrates on the surface, other hydrate particles may not be able to directly contact the ‘bare’ surface. Instead, they can adhere to the first gas hydrate layer, leading to an increased thickness of the hydrate deposit and a reduction in the flow cross-sectional area18. Therefore, hydrate adhesion with excess water coated on the CS surface was performed to simulate the scenario of a thick gas hydrate layer depositing on the flowline wall. Instead of using the water condensation method, where a thin water layer formed on the CS surface, a droplet was directly deposited and spread on the surface yielding a thick


Page 26 of 39

Page 27 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

water layer. The amount of water coated on the surface was 20 mg, leading to 0.6 ± 0.1 mg mm-2. A measured force at 2.07 MPa and 274 K was 17.9 ± 2.3 mN m-1 with a 10 second contact time between the gas hydrate particle and gas hydrate layer coating on the sample. In contrast, the hydrate adhesive force was 3.7 ± 0.7 mN m-1 in the CH4/C2H6 gas phase using the water condensation method (cf. ref.12), indicating that the increase of the gas hydrate layer can not only increase the adhesive force between the hydrate particle and surface, but also lead to larger error. To mimic gas hydrate deposition under continuous flow or during transient shut-in/restart operations in the flowline, which would be time dependent processes, the effect of contact time was performed to further investigate these deposition processes. Experimental results show that after 1 hour of contact, the hydrate particle was sintered to the gas hydrate deposit on the surface, and therefore no force could be measured due to the limitation (150 mN m-1) of the current apparatus. This time-dependent experiment indicates significant increases of the adhesive force when the contact time is increased. The possible mechanism of this observation is shown in Figure 12. Due to the presence of free water in the flowline, a gas hydrate particle can interact with deposits through the liquid bridge, which can dominate the interactions. After a long contact time, the liquid bridge converts to solid gas hydrates due to the dominant sintering process. With the increasing amount of gas hydrates repeating this mechanism, the thickness of the hydrate deposit increases, eventually resulting in a reduction of flow cross- sectional area.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 39

Figure 12. Possible mechanism of the gas hydrate adhesion process to the flowline wall, where blue represents a water phase, green a gas phase, brown a liquid hydrocarbon phase, and white for gas hydrates: (a) gas hydrate particle formed in the bulk phase with gas and water, where some gas hydrate deposits formed on the flowline wall; (b) some water converted to gas hydrate due to the contact with the gas hydrate particle; (c) more water converted to gas hydrate resulting in a buildup of the hydrate deposit on the pipe surface. Capillary bridge theory Comparison between experimental and calculated results According to the previous work10, 13, with the assumption of a negligible contribution of specific free energy, the final form of the capillary bridge model to estimate the interaction between two spheres can be given by equation (1):

2 cos ∗ ! " 30% 1 2/, ) &' ( ! * 30% 2 sin sin ∗

(1)

where FA is the cohesive force, γ is the interfacial tension of the bulk liquid, θp is the contact angle between gas hydrate particle and the liquid bridge, α is the embracing angle, H is particle separation distance, dsp/sp(H,V) is the liquid bridge immersion depth, and R* is the harmonic mean radius of two gas hydrate particles. These parameters are all also 27 ACS Paragon Plus Environment

Page 29 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

described in Figure 2. When the contact time is longer than 30 seconds, the bottom part of equation (1) was applied, where τt is the hydrate tensile strength, which is 0.91 MPa from Aman et al.,10 χ is the radius of contact. Based on equation (1), models for the volume (V) and immersion depth (dsp/sp(H,V)) of the capillary bridge theory can be given, as shown in equations (2) and (3): V πR) ) 0.5πR0 1

d

26 3−1 51 7 2 )

(2)

(3)

For the cohesive force of CH4/C2H6 gas hydrate particles in a model liquid hydrocarbon phase, it is expected that the change in interfacial tension of the water bridge in liquid hydrocarbon will be different compared to that in the vapor phase. Previous work suggested that the hydrate surface may be similarly energetic to that of water5, 33-36 when exposed to a liquid hydrocarbon; therefore, the interfacial tension of the water-bulk phase was applied in the interfacial calculations. For pure water in the vapor phase, the interfacial tension is 75.5 mN m-1 at 274 K37, and it decreases to 53.3 ± 0.5 mN m-1 when adding the model liquid hydrocarbon as a bulk liquid phase.9 Contact angle is another factor that has a strong effect on the final calculated results. Hireataa et al.38 have indicated that clathrate hydrates are hydrophilic through their experiments. The contact angle between a water droplet and a hydrophilic surface was reported as 40° by Sghaier and co-worker.39 Nevertheless, this experiment was perfomed with air as the bulk phase, whereas a liquid hydrocarbon served as the bulk phase for this present work. To understand the contact angle for a liquid hydrocarbon-water-surface system, Grate et al.40 reported a contact angle correlation between oil-water and air-water systems on


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 39

several different surfaces, as shown in Figure 13. Based on this correlation, a relationship was presented between the two systems. Therefore, the contact angle of an oil-watersurface system can be calculated using this correlation, yielding a final value of 55°.

Figure 13. Correlation of the cosines of the oil-water and air-water contact angles from Grate et al.40. Each point in this figure represents the contact angle measurements on silica and silanized silica surfaces in both air-water and oil-water systems. All these data points were generated from Grate et al.40. The resulting data fit is y = 0.667x + 0.384 (R = 0.981, n = 13). The cohesive force between two gas hydrate particles in a model liquid hydrocarbon phase was calculated by applying equation (1), all parameters were kept constant except interfacial tension and contact angle, which were previously discussed. These parameters were obtained from previous research.10,

13, 31, 39

The calculated cohesive force in the


Page 31 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

model liquid hydrocarbon phase was 25.9 mN m-1, which was close to our experimental value and they both were in the same order of magnitude (shown in Figure 14). This indicates the model predicted quite well the experimental data. Both the interfacial tension and contact angle discussed above present non-negligible contributions to both experiments and models, and they can also govern the gas hydrate cohesion and adhesion. By utilizing the same calculation method, the cohesive force calculated for systems with CH4/C2H6 gas mixture at 3.45 MPa12, CyC5 at atmosphere pressure10, and CyC5 hydrate with nitrogen at atmosphere pressure10 were also presented. It was found the liquid bridge model performed better for systems with a liquid phase, while a relatively large deviation was found when exposed to a vapor phase (see Figure 14). The possible explanations for this behavior could be the decrease of contact angle and an increase of the quasi-liquid layer. Further investigations are necessary to determine the contributions of these two


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

parameters.

Figure 14. Comparison between calculated (■) and experimental (□) cohesive force results. Four systems are presented in this graph : (a) gas hydrate particles with liquid hydrocarbon as the continuous phase; (b) gas hydrate particles withCH4/C2H6 gas mixture12 as the continuous phase; (c) CyC5 hydrate particles with CyC5 as continuous phase10; (d) CyC5 hydrate particles with N2 as the continuous phase10. The subcooling for these experiments are (from (a) to (d)): 10K, 6K, 4K and 4K, respectively. As contact time increases to above 30 seconds, the bottom part of equation (1) is applied, where the force equals to the product of the tensile strength of the hydrate particle and the contact area. From previous work, the radius of contact area can be calculated through equation (4):10 31 ACS Paragon Plus Environment

Page 32 of 39

Page 33 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

χ 1.2147 ∙ !

(5)

Based on a slight modification of this model (i.e. replacing the fitting parameter from 1.2147 to 1.95 for the gas hydrate system, rather than the CyC5 hydrate system), cohesive forces for different contact times were predicted and compared against the experimental and fitted results (see Figure 9, blue line). The calculated and measured results showed reasonable agreement. It should be noted the model proposed in from Aman et al.10 was generated and fitted from the experimental data using CyC5 hydrates. Therefore, equation (4) was modified by fitting to the gas hydrate experimental data. When the contact time increases to 12 hours, the measured force exceeded the limitation of the current fiber, which was 150 mN m-1. The predictions using the above model showed close agreement with the experimental data (as shown in Figure 9). Comparison between interparticle and particle-surface interactions for gas hydrates and CyC5 hydrates Figure 15 illustrates the comparison of hydrate cohesive and adhesive forces in different systems, including gas hydrate cohesion in a CH4/C2H6 gas mixture12, gas hydrate adhesion in a CH4/C2H6 gas mixture12, gas hydrate cohesion in a model liquid hydrocarbon, gas hydrate adhesion in a model liquid hydrocarbon, CyC5 hydrates in nitrogen at atmosphere pressure10, CyC5 hydrates liquid CyC5 at atmosphere pressure10. By comparing these results, experiments with gas hydrate showed interaction cohesive/adhesive forces that were one order of magnitude higher than the system with CyC5. For both gas and liquid phase measurements using gas hydrates, the cohesive forces were one order of magnitude higher than the adhesive forces. This figure further illustrates the difference between gas and liquid bulk phases. This could be explained by 32 ACS Paragon Plus Environment

Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 39

the change of interfacial tension and contact angle therefore changing the surface free energy of hydrate particles. The surface free energy can be calculated by applying equation (6)14, 41-44 as shown below: cos −1 25

? 1 − A= ? − @? ) @?

(6)

where the θP represents contact angle, γsv is the surface free energy (also known as the solid/vapor interfacial tension), γlv is the liquid/vapor interfacial tension, β1 is a constant equal to 0.0001057 mJ/m2. When the hydrate particles are exposed to a liquid hydrocarbon phase, the vapor changed to liquid hydrocarbon, therefore the γlv is the interfacial tension between water and liquid hydrocarbon. When the gas mixture acts as the bulk phase, γlv is the liquid/gas interfacial tension. Both γlv and θP change due to the conversion of the bulk phase. Based on this method, the surface free energy of hydrates in the liquid hydrocarbon -dominated system is calculated to be 35.5 mN m-1, whereas it increases to 58.9 mN m-1 for the gas-dominated system. Previous work by Aspenes et al.14 showed an increase in the interactions with increases in the surface free energy. Therefore, from the surface free energy perspective, the latter calculation in this study may help to provide insight into why a stronger interaction was observed in the gas phase compared to the liquid hydrocarbon phase. In all these different systems, interfacial phenomena govern the interactions between particle-particle and particle-surface systems, therefore, further systematic investigations are required to provide further insight into the contribution of different interfacial parameters.


Page 35 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

Figure 15. Comparison of hydrate cohesive/adhesive forces in different systems: gas hydrate cohesive force in CH4/C2H6 gas mixture12, gas hydrate adhesive force in CH4/C2H6 gas mixture12, gas hydrate cohesive force in model liquid hydrocarbon, gas hydrate adhesive force in model liquid hydrocarbon, CyC5 hydrates cohesive force in nitrogen at atmosphere pressure10, CyC5 hydrates cohesive force liquid CyC5 at atmosphere pressure10, respectively. L: liquid hydrocarbon bulk phase; V: vapor bulk phase.

CONCLUSIONS We present the first direct measurements on the cohesive and adhesive force studies of CH4/C2H6 gas hydrate in a liquid hydrocarbon-dominated system utilizing a high pressure


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

micromechanical force (HP-MMF) apparatus. A CH4/C2H6 gas mixture was used as gas hydrate former in the model liquid hydrocarbon phase. For the cohesive force baseline tests, it was found that the addition of hydrocarbon can change the interfacial tension and contact angle of water in liquid hydrocarbon compared to water in the gas phase, resulting in a force of 23.5 ± 2.5mN m-1 at 3.45 MPa, 274 K and 2-hour annealing time. It was observed that the cohesive force was inversely proportional to the annealing (hydrate shell growth) time. By comparing the cohesive force at high pressure and atmosphere pressure, although they showed similar trends in behavior with longer annealing time, the cohesive force at high pressure with a liquid hydrocarbon was one order of magnitude higher than that at ambient pressure. The hydrate interparticle force was found to increase with increasing the contact time. For longer contact time (> 12 hour), the force exceeded the limits of the measurement, because the two hydrate particles adhered permanently to form one large particle. The particle-surface adhesive force in the model liquid hydrocarbon was measured to be 5.3 ± 1.1 mN m-1 at the same experimental condition as interparticle measurement in a liquid hydrocarbon phase. Finally, with 1-hour contact time, the hydrate particle and the carbon steel (CS) surface were sintered together and the force (>150 mN m-1) was higher than what could be measured by the current apparatus. The calculated result using the capillary bridge model was found to be in close agreement with the experimental results. The results from a gas-dominated system was higher than that from a liquid hydrocarbondominated system, mainly likely due to the change in the surface free energy which is governed by the interfacial tension and contact angle. These results illustrate the


Page 36 of 39

Page 37 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

interfacial mechanisms can change with changing the bulk phases; they also governed the interparticle and particle-surface interactions.

ACKNOWLEDGEMENTS We would like to acknowledge funding and support from the CSM Center for Hydrate Research Consortium; William K. Coors Distinguished Chair Fund; Oceanit for providing the surface samples; all the hydratebusters for their help and advice.

REFERENCES 1. Mahajan, D.; Taylor, C. E.; Mansoori, G. A., An introduction to natural gas hydrate/clathrate: The major organic carbon reserve of the Earth. Journal of Petroleum Science and Engineering 2007, 56 (1-3), 1-8. 2. Ramaswamy, D.; Sharma, M. M. The effect of surfactants on the kinetics of hydrate formation. Proc. Society of Petroleum Engineers' Annual Technical Conference and Exhibition (ATCE), SPE-141226-MS, TX, 2011. 3. Sloan, E. D.; Koh, C. A., Clathrate hydrates of natural gases. 3rd Edition; CRC Press, Taylor & Francis Group, Boca Raton, FL, 2007. 4. Hammerschmidt, E. G., Formation of gas hydrates in natural gas transmission lines. Industrial & Engineering Chemistry 1934, 26 (8), 851-855. 5. Aman, Z. M.; Koh, C. A., Interfacial phenomena in gas hydrate systems. Chemistry Society Reviews 2016, 45 (6), 1678-90. 6. Fowler, R. H., A tentative statistical theory of macleod's equation for surface tension, and the parachor. Proc. Royal Society of London. Series A - Mathematical and Physical Sciences 1937, 159 (897), 229-246. 7. Fowler, R. H.; Guggenheim, E. A., Statistical thermodynamics: a version of statistical mechanics for students of physics and chemistry. Cambridge University Press, The Macmillan Company, 1939. 8. Israelachvili, J. N., Intermolecular & surface forces. 2nd Edition; Academic Press, Elsevier Inc., 1991. 9. Brown, E. Studies of hydrate cohesion, adhesion and interfacial properties using micromechanical force measurements. Ph.D. Thesis, Colorado School of Mines, Golden, CO, 2016. 10. Aman, Z. M.; Brown, E. P.; Sloan, E. D.; Sum, A. K.; Koh, C. A., Interfacial mechanisms governing cyclopentane clathrate hydrate adhesion/cohesion. Physical Chemistry Chemical Physics 2011, 13 (44), 19796-19806. 11. Yang, S.; Kleehammer, D. M.; Huo, Z.; Sloan, E. D.; Miller, K. T., Temperature dependence of particle–particle adherence forces in ice and clathrate hydrates. Journal of Colloid and Interface Science 2004, 277 (2), 335-341.


Langmuir

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

12. Wang, S.; Hu, S.; Brown, E. P.; Nakatsuka, M. A.; Zhao, J.; Yang, M.; Song, Y.; Koh, C. A., High pressure micromechanical force measurements of the effects of surface corrosion and salinity on CH4/C2H6 hydrate particle-surface interactions. Physical Chemistry Chemical Physics 2017, 19 (20), 13307-13315. 13. Rabinovich, Y. I.; Esayanur, M. S.; Moudgil, B. M., Capillary forces between two spheres with a fixed volume liquid bridge: theory and experiment. Langmuir 2005, 21 (24), 10992-10997. 14. Aspenes, G. The onfluence of pipeline wettability and crude oil composition on deposition of gas hydrates during petroleum production. Ph.D. Thesis, University of Bergen, Bergen, 2009. 15. Dieker, L.E. Cyclopentane hydrate interparticle adhesion force measurements. MS. Thesis, Colorado School of Mines, Golden, CO, 2009. 16. Smith, J. D.; Meuler, A. J.; Bralower, H. L.; Venkatesan, R.; Subramanian, S.; Cohen, R. E.; McKinley, G. H.; Varanasi, K. K., Hydrate-phobic surfaces: fundamental studies in clathrate hydrate adhesion reduction. Physical Chemistry Chemical Physics 2012, 14 (17), 6013-6020. 17. Aman, Z. M.; Leith, W. J.; Grasso, G. A.; Sloan, E. D.; Sum, A. K.; Koh, C. A., Adhesion force between cyclopentane hydrate and mineral surfaces. Langmuir 2013, 29 (50), 15551-7. 18. Brown, E.; Hu, S.; Wang, S.; Wells, J.; Nakatsuka, M.; Veedu, V.; Koh, C. A., Lowadhesion coatings as a novel gas hydrate mitigation strategy. Proc. Offshore Technology Conference, OTC-27874-MS, 2017. 19. Jung, J. W.; Santamarina, J. C., Hydrate adhesive and tensile strengths. Geochemistry, Geophysics, Geosystems 2011, 12 (8), 1-9. 20. Song, J. H.; Couzis, A.; Lee, J. W., Direct measurements of contact force between clathrate hydrates and water. Langmuir 2010, 26 (12), 9187-9190. 21. Lee, W.; Baek, S.; Kim, J.-D.; Lee, J. W., Effects of salt on the crystal growth and adhesion force of clathrate hydrates. Energy & Fuels 2015, 29 (7), 4245-4254. 22. Lee, B. R.; Sum, A. K., Micromechanical cohesion force between gas hydrate particles measured under high pressure and low temperature conditions. Langmuir 2015, 31 (13), 38843888. 23. Majid, A. A. A. An investigation on the viscosity and transportability of methane hydrate slurries using a high pressure rheometer and flowloop. Ph.D. Thesis, Colorado School of Mines, Golden, CO, 2015. 24. Bienfait, M., Roughening and surface melting transitions: consequences on crystal growth. Surface Science 1992, 272 (1), 1-9. 25. Pavlovska, A.; Dobrev, D.; Bauer, E., Orientation dependence of the quasi-liquid layer on tin and indium crystals. Surface Science 1994, 314 (3), 341-352. 26. Döppenschmidt, A.; Butt, H.-J., Measuring the thickness of the liquid-like layer on ice surfaces with atomic force microscopy. Langmuir 2000, 16 (16), 6709-6714. 27. Johnson, K. L.; Kendall, K.; Roberts, A. D., Surface energy and the contact of elastic solids. Proc. Royal Society of London. A. Mathematical and Physical Sciences 1971, 324 (1558), 301-313. 28. Derjaguin, B. V.; Muller, V. M.; Toporov, Y. P., Effect of contact deformations on the adhesion of particles. Journal of Colloid and Interface Science 1975, 53 (2), 314-326. 29. Willett, C. D.; Adams, M. J.; Johnson, S. A.; Seville, J. P. K., Capillary Bridges between Two Spherical Bodies. Langmuir 2000, 16 (24), 9396-9405. 30. Schubert, H., Capillary forces - modeling and application in particulate technology. Powder Technology 1984, 37 (1), 105-116. 31. Rabinovich, Y. I.; Adler, J. J.; Esayanur, M. S.; Ata, A.; Singh, R. K.; Moudgil, B. M., Capillary forces between surfaces with nanoscale roughness. Advances in Colloid and Interface Science 2002, 96 (1), 213-230.


Page 38 of 39

Page 39 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Langmuir

32. Kruis, F. E.; Kusters, K. A.; Pratsinis, S. E.; Scarlett, B., A simple model for the evolution of the characteristics of aggregate particles undergoing coagulation and sintering. Aerosol Science and Technology 1993, 19 (4), 514-526. 33. Kwok, D. Y.; Neumann, A. W., Contact angle measurement and contact angle interpretation. Advances in Colloid and Interface Science 1999, 81 (3), 167-249. 34. Kwok, D. Y.; Neumann, A. W., Contact angle interpretation in terms of solid surface tension. Colloids and Surfaces A: Physicochemical and Engineering Aspects 2000, 161 (1), 31-48. 35. Asserson, R. B.; Hoffmann, A. C.; Høiland, S.; Asvik, K. M., Interfacial tension measurement of freon hydrates by droplet deposition and contact angle measurements. Journal of Petroleum Science and Engineering 2009, 68 (3), 209-217. 36. Goebel, A.; Lunkenheimer, K., Interfacial Tension of the Water/n-Alkane Interface. Langmuir 1997, 13 (2), 369-372. 37. Petrova, T., Revised release on surface tension of ordinary water substance; The International Association for the Properties of Water and Steam Moscow, Russia, 2014. 38. Hirata, A.; Mori, Y. H., How liquids wet clathrate hydrates. Chemical Engineering Science 1998, 53 (14), 2641-2643. 39. Sghaier, N.; Prat, M.; Ben Nasrallah, S., On the influence of sodium chloride concentration on equilibrium contact angle. Chemical Engineering Journal 2006, 122 (1), 47-53. 40. Grate, J. W.; Dehoff, K. J.; Warner, M. G.; Pittman, J. W.; Wietsma, T. W.; Zhang, C.; Oostrom, M., Correlation of oil–water and air–water contact angles of diverse silanized surfaces and relationship to fluid interfacial tensions. Langmuir 2012, 28 (18), 7182-7188. 41. Neumann, A. W.; Good, R. J.; Hope, C. J.; Sejpal, M., An equation-of-state approach to determine surface tensions of low-energy solids from contact angles. Journal of Colloid and Interface Science 1974, 49 (2), 291-304. 42. Li, D.; Neumann, A. W., A reformulation of the equation of state for interfacial tensions. Journal of Colloid and Interface Science 1990, 137 (1), 304-307. 43. Li, D.; Neumann, A. W., Contact angles on hydrophobic solid surfaces and their interpretation. Journal of Colloid and Interface Science 1992, 148 (1), 190-200. 44. Tavana, H.; Neumann, A. W., Recent progress in the determination of solid surface tensions from contact angles. Advances in Colloid and Interface Science 2007, 132 (1), 1-32.


of 39 ACS Paragon Plus Environment Langmuir 1 ... - ACS Publications

Recommend Documents